© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 15078
    Prev
    Next
    INDEX
    Explanations

    names of a specific person, "Kurt"

    oai_token-act-pair · gpt-3.5-turbo

    references to individuals named Kurt and Courtney

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PC
    -0.69
     si
    -0.66
     gui
    -0.65
     signalling
    -0.65
    astern
    -0.64
     CR
    -0.63
     âĶĤ
    -0.62
    eyes
    -0.62
     maid
    -0.60
     poultry
    -0.60
    POSITIVE LOGITS
     Kurt
    3.73
     Nirvana
    1.36
     Courtney
    1.32
     Von
    1.22
     Bret
    1.18
     Curtis
    1.18
     Holt
    1.16
     Klaus
    1.15
     Hulk
    1.14
     Curt
    1.12
    Activations Density 0.021%

    No Known Activations